京都大学 / 0048 / 新制・課程博士 / 博士(農学) / 甲第21159号 / 農博第2285号 / 新制||農||1060(附属図書館) / 学位論文||H30||N5133(農学部図書室) / 京都大学大学院農学研究科応用生命科学専攻 / (主査)教授 植田 和光, 教授 阪井 康能, 教授 矢﨑 一史 / 学位規則第4条第1項該当 / Doctor of Agricultural Science / Kyoto University / DFAM

No description available.

This study was motivated by a theoretical formulation on mechanobiology of soft and hard skeletal tissue differentiation. To prove this formulation experimentally, I hypothesized that cartilaginous phenotype can be induced in vitro in a seemingly non-cartilaginous cell source from fibrous tissue. In testing this hypothesis, I have focused on cartilage as a target and fibrous tissue as an origin or the source of cell. Four different trials were pursued with one supposition in common, i.e. hydrostatic pressure is one of the main driving forces for chondroinduction in vitro. The first and second trials pertained to the influence of a relatively short and long duration cyclic hydrostatic compression on rat Achilles tendon fibroblasts. The third trial was to examine the effect of two different drugs on cytoskeletal elements of mesenchymal stem cells or mouse embryonic fibroblast lines in pellet cultures combined with the similar duration and/or frequency of cyclic hydrostatic pressure adopted in the aforesaid trials with no pharmacological agents added. Last, attempts were made to implement an advanced technique in molecular biology called 'PCR array' to further quantify expression levels of eighty four pathway-specific genes in mouse TGFbeta/BMP signaling traffic under the same physiological regimen of hydrostatic compression. Results demonstrated that transdifferentation in phenotype from tendon to fibrocartilage may have occurred in vitro in tendon fibroblasts in pellet cultures exposed to hydrostatic pressure. Experiments on the role of the cytoskeleton in mechanotransduction of the applied level of hydrostatic pressure demonstrated that disruption of microfilaments in the presence of cytochalasin-D did not significantly interfere with the anabolic effect of cyclic pressure. However, disruption of microtubule assembly by nocodazole abolished the pressure-induced stimulation in cartilage marker genes. These findings suggest that microtubules, but not microfilaments, are involved in mechanotransduction of hydrostatic pressure by mesenchymal stem cells.

No description available.

No description available.

No description available.

No description available.

No description available.

Partial or complete tears of the anterior cruciate ligament (ACL) can greatly afflict quality of life and often require surgical reconstruction with autograft or allograft tissue to restore native knee biomechanical function. However, limitations exist with these treatments that include donor site pain and weakness found with autografts, and longer "ligamentization" and integration times due to the devitalization of allograft tissue. Alternatively, a tissue engineering approach has been proposed for the fabrication of patient-specific grafts that can more rapidly and completely heal after ACL reconstruction. Electrospun micro-fiber networks have been widely utilized as biomaterial scaffolds to support the growth and differentiation of mesenchymal stem cells toward many tissue lineages including ligament. However, these micro-fiber networks do not possess suitable sizes and shapes for a ligament application and cannot support cell infiltration. The objective of this work was to develop techniques to 1) rapidly cellularize micro-fiber networks, 2) assemble micro-fiber networks into cylindrical composites, 3) provide cues to mesenchymal stem cells (MSCs) to guide their differentiation toward a ligament phenotype. The cellularization of micro-fiber networks was performed utilizing a co-electrospinning/electrospraying technique. Cells deposited within a cell culture medium solution remained where they were deposited and did not proliferate. The inclusion of space-filling hydrogel network such as collagen was necessary to reduce the density of the micro-fiber network to facilitate spreading. However, it became apparent that the incorporation of significant collagen phase was necessary for long-term MSC survival within the micro-fiber network. Next, two approaches were developed to fabricate large cylindrical, composites. The first approach utilized a co-electrospinning/electrospraying technique to generate micro-fiber/collagen composites that were subsequently rolled into cylinders. These cylindrical composites exhibited greater diameters and water weight percentages as collagen content increased. However, the high micro-fiber content of these composites was inhibitory to cell survival. In the second approach, thin layers (~5-10 fibers) of aligned electrospun PEUR fibers were encapsulated within a collagen gel and subsequently rolled the composites into cylinders. These sparse-fiber composites were nearly 98% by weight water and confocal imaging revealed the presence of sparse fiber layers (~5 fibers thick) separated by approximately 200 μm thick collagen layers. We hypothesize that the proliferation and migration of MSCs within these micro-fiber/collagen composites may not be restricted by the presence of a dense, non-manipulatable electrospun fiber network present in traditionally rolled fiber composites. Simple model platforms were then developed to study the influence of sparse micro-fibers on MSCs differentiation within a collagen hydrogel. MSCs in the presence of the softest (5.6 MPa) micro-fibers elongated and oriented to the underlying network and exhibited greater expression of scleraxis, and α-smooth muscle actin compared to the stiffest (31 MPa) fibers. Additionally, preliminary results revealed that the incorporation of fibroblast growth factor-2 and growth and differentiation factor-5 onto micro-fibers through chemical conjugation enhanced expression of the ligamentous markers collagen I, scleraxis, and tenomodulin. In conclusion, micro-fiber/collagen composite materials must possess sufficient space to support the infiltration and differentiation of MSCs. The strategies described in this document could be combined to fabricate large, micro-fiber/collagen composites that can support cell infiltration and provide relevant cues to guide the formation of an engineered ligament tissue. / Ph. D.

<p dir="ltr">Tissue engineering utilizes polymers, cells, and other bioactive factors to promote regeneration within damaged tissue. The main works in this thesis employ naturally derived polymers for use in tissue engineering and explores ways to recapitulate native environments <i>in vitro</i>.</p><p dir="ltr">Collagen (col) is the most prevalent protein in the body. Col type I, II, and III are all fibril-forming collagens that provide structure to tissues. All three types polymerize <i>in vitro</i> to form hydrogels, and these hydrogels have often been studied for use in tissue engineering. Other applications include <i>in vitro </i>tissue models for studies on drug diffusion and drug delivery. Blending collagen types is of particular interest as col I is easier to source and is therefore cheaper than other collagen types. However, to confer biological signals to tissues where col II or III are more abundant (e.g., cartilage or cardiac tissue, respectively), col II or III can be added to col I to form col I/II or col I/III gels, respectively. Additionally, adding multiple types of col to hydrogel models better recapitulates the native environment and can better capture effects on drug diffusion. In this work, compared to col I alone, col I/II hydrogels polymerize more slowly, form more fibril bundles, result in softer hydrogels, and impede transport of larger macromolecules. On the other hand, col I/III gels polymerize at a similar rate to col I, create heterogenous fibril structures, are oftentimes stiffer than col I, and also impede transport of larger macromolecules. Additionally, this work explored the effect of polymerization temperature on blended gel polymerization and properties.</p><p dir="ltr">The second work evaluates col I/II hydrogels for a specific application: cartilage tissue engineering for osteoarthritic applications. Col II is the primary protein found in cartilage. Other components include: glycosaminoglycans, such as hyaluronic acid (HA) and chondroitin sulfate, chondrocytes (cartilage cells), and other small signaling molecules. Building on prior work in the group, high molecular weight hyaluronic acid (HA) was added to col I/II hydrogels, and cartilage differentiation of mesenchymal stem cells (MSCs) was assessed under ideal laboratory conditions and under pro-inflammatory, osteoarthritic conditions (i.e., cytokine-supplemented media of oncostatin M (OSM) at 10 ng/mL and tumor necrosis factor-α (TNF-α) at 20 ng/mL). The addition of HA did not dramatically impact cartilage differentiation of MSCs, however, HA did mitigate the effect of inflammation via downregulation of a degradative enzyme. HA had little impact on inflammatory cytokine production of interleukin (IL)-6 or IL-8, both of which are upregulated during osteoarthritis. However, a linear model suggests that HA and IL-8 are strongly correlated. Thus, this system should be explored further with different HA concentrations or presentations (e.g., chemically modified).</p><p dir="ltr">The last primary chapter of this thesis provides depth to the pro-inflammatory, osteoarthritic model used in the previous chapter. Different pro-inflammatory environments are studied using cytokines found in OA. MSC pellets (used in literature as controls to confirm chondrogenic potential of MSCs) were used to evaluate these inflammatory environments since MSCs are commonly used in tissue engineering. Six treatments were studied: negative control (without the chondrogenic growth factor TGF-β3), positive control (with the chondrogenic growth factor TGF-β3), and four cytokine treatments all with TGF-β3. First, IL-1β at 10 ng/mL was utilized as a comparison to literature. The other three cytokine groups used TNF-α at 20 ng/mL and OSM at 10 ng/mL individually or combined to form the main experimental group, OSM+TNF-α. All cytokine treatment groups limited cartilage production, but OSM decreased production to a statistically lesser extent than other cytokine groups. This trend was similar to observations made via immunostaining of cartilage matrix and gene expression analysis of aggrecan. Furthermore, OSM+TNF-α statistically lowered aggrecan gene expression. In terms of degradation, when compared to all other groups, OSM dramatically increased the protein expression of the degradative enzyme matrix metalloproteinase-13 (MMP-13). Evaluation of inflammatory markers (IL-6 and IL-8) revealed no signal for OSM-treated pellets. TNF-α yielded some signal after 1 week in culture but no signal after two weeks. IL-1β and OSM+TNF-α both resulted in sustained IL-6 and IL-8 expression, however, IL-1β exhibited large variance. Thus, each cytokine contributes to various pathways that are present in OA. Since the combination of OSM and TNF-α appeared to lower cartilage gene expression and resulted in sustained and reproducible IL-6 and IL-8 production, it may serve as a better model of OA than a single cytokine such as IL-1β.</p>